Re: [PATCH] ext4: Ensure zeroout blocks have no dirty metadata

[Curt Wohlgemuth <curtw@google.com>]

Hi Ted:

On Tue, Dec 29, 2009 at 10:53 AM,  <tytso@mit.edu> wrote:
> On Tue, Dec 29, 2009 at 09:56:29AM -0800, Curt Wohlgemuth wrote:
>>
>> Thanks for pointing out the arithmetic problems in my patch.  Yours
>> below looks good to me.
>
> Hi Curt,
>
> Can you do me a favor and send out a full patch with Aneesh's
> correction?

Here it is.  This is against 2.6.33-rc2, if that makes a difference.  Thanks!

====================================================================

This fixes a bug in which new blocks returned from an extent created with
ext4_ext_zeroout() can have dirty metadata still associated with them.

This patch was originally by Aneesh Kumar <aneesh.kumar@linux.vnet.ibm.com>.

       Signed-off-by: Curt Wohlgemuth <curtw@google.com>

---


diff -uprN orig/fs/ext4/extents.c new/fs/ext4/extents.c
--- orig/fs/ext4/extents.c	2009-12-29 14:59:12.000000000 -0800
+++ new/fs/ext4/extents.c	2009-12-29 15:18:43.000000000 -0800
@@ -3023,6 +3023,14 @@ out:
 	return err;
 }

+static void unmap_underlying_metadata_blocks(struct block_device *bdev,
+			sector_t block, int count)
+{
+	int i;
+	for (i = 0; i < count; i++)
+                unmap_underlying_metadata(bdev, block + i);
+}
+
 static int
 ext4_ext_handle_uninitialized_extents(handle_t *handle, struct inode *inode,
 			ext4_lblk_t iblock, unsigned int max_blocks,
@@ -3098,6 +3106,18 @@ out:
 	} else
 		allocated = ret;
 	set_buffer_new(bh_result);
+	/*
+	 * if we allocated more blocks than requested
+	 * we need to make sure we unmap the extra block
+	 * allocated. The actual needed block will get
+	 * unmapped later when we find the buffer_head marked
+	 * new.
+	 */
+	if (allocated > max_blocks) {
+		unmap_underlying_metadata_blocks(inode->i_sb->s_bdev,
+					newblock + max_blocks,
+					allocated - max_blocks);
+	}
 map_out:
 	set_buffer_mapped(bh_result);
 out1:
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html